Upload
others
View
0
Download
0
Embed Size (px)
Citation preview
DISCUSSION PAPER SERIES
IZA DP No. 12254
Artjoms IvlevsRoswitha King
To Europe or Not to Europe? Migration and Public Support for Joining the European Union in the Western Balkans
MARCH 2019
Any opinions expressed in this paper are those of the author(s) and not those of IZA. Research published in this series may include views on policy, but IZA takes no institutional policy positions. The IZA research network is committed to the IZA Guiding Principles of Research Integrity.The IZA Institute of Labor Economics is an independent economic research institute that conducts research in labor economics and offers evidence-based policy advice on labor market issues. Supported by the Deutsche Post Foundation, IZA runs the world’s largest network of economists, whose research aims to provide answers to the global labor market challenges of our time. Our key objective is to build bridges between academic research, policymakers and society.IZA Discussion Papers often represent preliminary work and are circulated to encourage discussion. Citation of such a paper should account for its provisional character. A revised version may be available directly from the author.
Schaumburg-Lippe-Straße 5–953113 Bonn, Germany
Phone: +49-228-3894-0Email: [email protected] www.iza.org
IZA – Institute of Labor Economics
DISCUSSION PAPER SERIES
ISSN: 2365-9793
IZA DP No. 12254
To Europe or Not to Europe? Migration and Public Support for Joining the European Union in the Western Balkans
MARCH 2019
Artjoms IvlevsUniversity of the West of England and IZA
Roswitha KingØstfold University College
ABSTRACT
IZA DP No. 12254 MARCH 2019
To Europe or Not to Europe? Migration and Public Support for Joining the European Union in the Western Balkans
For decades, countries aspiring to join the European Union (EU) have been linked to it
through migration. Yet little is known about how migration affects individual support
for joining the EU in prospective member states. We explore the relationship between
migration and support for EU accession in the Western Balkans. Using data from the Gallup
Balkan Monitor survey, we find that prospective and return migrants, as well as people with
relatives abroad, are more likely to vote favourably in a hypothetical EU referendum. At the
same time, only people with relatives abroad are more likely to consider EU membership
a good thing. Our results suggest that migration affects attitudes toward joining the EU
principally through instrumental/utilitarian motives, with channels related to information
and cosmopolitanism playing only a minor role. Overall, our study suggests that migration
fosters support for joining a supranational organization in the migrants’ countries of origin,
which, in turn, is likely to affect political and institutional development of these countries.
JEL Classification: F22, F24, J61, 052
Keywords: migration, return migration, European Union, european integration, Western Balkans
Corresponding author:Artjoms IvlevsDepartment of AccountingEconomics and FinanceBristol Business SchoolUniversity of the West of EnglandBristol BS16 1QYUnited Kingdom
E-mail: [email protected]
2
1. INTRODUCTION
Migration has, for better or worse, played a key role in the process of European integration
(Azrout et al., 2011; Hobolt et al. 2011; Gerhards and Hans 2011, de Vreese et al. 2008;
McLaren 2007; Favell, 2011; Kuhn, 2015; Geddes, 2014; Lawless, 2015). Much of Europe’s
contemporary economic and social landscape has been shaped by rural, regional, and cross-
border labor migration dating back to pre-industrialization times (Favell, 2011). Today, the
right to move and work freely within Europe is one of the Common Market’s pillars, and
people engaged in transnational European Union (EU) mobility are more likely to support
European integration (Kuhn, 2015). Meanwhile, immigration from the New Member States
which joined the EU in 2004 and 2007 and, even more so, immigration from outside Europe
are increasing Euroscepticism,2 especially in ‘old’ member states. It is, for example, well
documented that an immigrant presence in one’s place of residence and unfavorable attitudes
toward immigration are strong determinants of opposition toward further EU enlargement
(Azrout et al., 2011; Hobolt et al. 2011; Gerhards and Hans 2011, de Vreese et al. 2008;
McLaren 2007).3 The 2016 EU referendum in the UK, where the majority voted to leave,
provides a vivid example of a negative effect of intra-EU migration on European integration.
Central to this referendum’s campaign was immigration from new EU member states, which,
for some time, had been fuelling anti-EU sentiment in the country (Geddes, 2014; Ivlevs and
Veliziotis, 2018; Lawless, 2015).
2 We define Euroscepticism as opposition to the EU, its policies, and further integration.
3 There is also a large, more general literature on immigration’s effects on anti-immigration sentiment of
receiving populations. See Pottie-Sherman and Wilkes (2017) for a summary of the evidence.
3
In this context, one question has been overlooked: How does migration affect public support
for joining the EU in prospective EU countries? This article sets out to answer this question,
which is important for several reasons. First, historically, prospective EU countries have been
linked with ‘old’ EU member states through migration (Brunnbauer, 2009), and migrants
might have provided insider knowledge about EU countries, thus influencing views about
joining the EU among those staying behind. For example, people from the successor states of
Yugoslavia actively participated in the post-WWII guestworker programs in the
industrialized countries of Western Europe. In 1971, one in five of all employed Yugoslavs
worked abroad, mainly in West Germany and Austria but also in France, Switzerland, and
Sweden (Brunnbauer, 2009). This migration, later followed by family reunification
(Zimmerman, 1987) and war refugee flows, has not only sustained livelihoods for thousands
of households but also been a source of information about the EU in the migrant origin
countries, potentially affecting willingness to join the EU. Second, public support for
European integration has a significant bearing on the EU’s democratic legitimacy (Gabel,
1998; Hobolt and De Vries, 2016; Recchi and Kuhn, 2013). Given current economic woes
and declining enthusiasm for further European integration and expansion inside the EU, it is
of interest whether the European project is still in high standing in the eyes of prospective EU
members and whether factors like migration could drive attitudes in these countries toward
joining the EU.
In this article, we study the relationship between migration and support for joining the EU in
the successor states of Yugoslavia (Bosnia and Herzegovina, Croatia, Former Yugoslav
Republic of Macedonia, Kosovo, Montenegro, Serbia) as well as Albania. These countries are
4
the next in line to join the EU4 and are characterized by a high incidence of migration, much
of it to the EU. Yet support for EU membership in the Western Balkan region is far from
universal and varies greatly within and between countries. For example, in Croatia’s
referendum on joining the EU in January 2012, one third of the participating population voted
against joining (BBC, 2012). In 2010, the share of people considering EU membership a
good thing varied from 87% in Kosovo, 81% in Albania, and 73% in Montenegro to 44% in
Serbia and 25% in Croatia (Gallup, 2010). Polls also reveal declining support for the EU
between 2010 and 2015 across the Western Balkan region (Belloni, 2016).
Several factors may have contributed to rising Euroscepticism in the Western Balkans. The
ongoing problems in the EU – economic hardships in neighboring Greece following the Euro
crisis, the 2015 refugee crisis, and the 2016 UK referendum on whether to stay in the EU –
may cast doubt on the EU’s ability to deliver peace and prosperity (Belloni and Brunazzi,
2017; Ker-Lindsay, 2017; Yamron, 2017). Some countries of ‘old’ Europe have recently
witnessed an increase in anti-immigration sentiment and a rise of far-right parties, in part
attributable to the realities of post-2004 immigration from Central and Eastern Europe
(Geddes, 2014; Halla, 2017). Such developments may make people in the Western Balkans
wonder whether they are welcome in the EU and contribute to their Euroscepticism (Ker-
Lindsay, 2017). Among internal factors, war and conflict memories may still be fresh in the
Western Balkans and predispose some people against joining the EU over resentment that the
EU took sides in the Balkan conflict. For example, parts of Serbia’s population likely harbour
4 Croatia joined the EU in 2013 but was not yet an EU member at the time the interview survey (2009-11) on
which this study is based. Slovenia, another successor state of Yugoslavia, joined the EU in 2004 and was not
included in the survey examined here.
5
negative feelings because many EU member states were among the first to recognize the
statehood of independent Kosovo, which formerly was part of Serbia (Nuttall, 2017; Belloni,
2016).
Against this background, we study whether migration-related variables – willingness to
emigrate, having relative(s) abroad, and being a return migrant – have constrained or, on the
contrary, propelled euro-optimism in the Western Balkan region. In Section 2, we
hypothesize that migration affects support for joining the EU through a combination of
instrumental motives and making residents better informed about the EU. We are also
interested in whether migration in general, with its presumed ‘cosmopolitan effect,’ or only
migration specifically to the EU shapes public attitudes toward joining the bloc. We use data
from a large and underexplored Gallup Balkan Monitor survey, described in Section 3, that
contain rich information about both migration and EU preferences and allow a nuanced
analysis of migration’s effect on the sentiment toward joining the EU.
Our empirical analysis, presented in Section 4, reveals that migration-related variables are
important determinants of support for EU membership among prospective members of the
Western Balkans. People considering emigration, people with relatives in an EU country, and
return migrants who have worked or studied in the EU are all more likely to vote favorably in
a hypothetical EU referendum. However, positive sentiment is not necessarily shown toward
the more general consideration of whether EU membership would be a good thing. As we
discuss in Section 5, these results suggest that migration affects attitudes toward joining the
EU mainly through instrumental motives, while informational and cosmopolitanism-related
channels receive little support.
6
In developing these ideas, this article contributes to two strands of literature. First, we add to
the literature on migration’s influence on support for European integration that has so far
focused on ‘old’ EU member states, within-EU mobility, and migration from outside Europe
(Azrout et al., 2011; De Vreese et al. 2008; Hobolt et al. 2011; Gerhards and Hans, 2011;
McLaren, 2007; Kuhn, 2015). The perspective of countries aspiring to join the EU remains
unexplored, and we fill this gap by focusing on the links between emigration and support for
joining the EU in the Western Balkans. Second, we contribute to the literature on
international migration’s effects on the political and institutional development in migrants’
origin countries (Barsbai et al., 2018; Batista and Vicente, 2011; Chauvet and Mercier, 2014;
Docquier et al., 2016; Ivlevs and King, 2017; Pfutze, 2012). We add to this body of
knowledge by showing that migration fosters support for joining a supranational institution
(the EU). As it has been argued and shown that joining the EU contributes to the democratic
and institutional reforms in candidate countries (Elbasani, 2013; Pop-Eleches, 2007; Pridham,
2005; Schimmelfennig, 2007; Schimmelfennig and Sedelmeier, 2008), it is likely that
migration will foster the political and institutional development in the successor states of
former Yugoslavia – through increased support for joining the EU.
2. MIGRATION AND SUPPORT FOR EU MEMBERSHIP: THEORETICAL
CONSIDERATIONS
This section presents an analytical framework for examining if and how attitudes toward
joining the EU might be influenced by migration-related considerations and experiences,
especially willingness to migrate, having relatives abroad, and being a return migrant. We
start by outlining the main explanations for public support of European integration (we
7
consider joining the EU as one dimension of European integration) and then discuss how
migration-related phenomena fit into these conceptual frameworks.
2.1. Main explanations of public support for European integration
The literature on the determinants of support for European integration has distinguished
between and among three core explanations for why people would support European
integration: 1) utilitarian, 2) identity, and 3) cue-taking (see Hobolt and De Vries, 2016). The
utilitarian approach posits that people will support EU integration if they see a personal
(egocentric) or collective (sociotropic) benefit from deeper integration. It is often assumed
that at the individual level, people with higher levels of human and financial capital –
education, income, and occupational skills – benefit most from European integration as they
better adapt to competition and are more likely to take advantage of the opportunities offered
by a liberalized European market (Anderson and Reichert, 1995; Gabel and Palmer, 1995;
Inglehart, 1970; Tucker et al., 2002). At the same time, low-skilled residents lose out as
deeper integration leads to production shifting across borders and more job insecurity (Gabel
and Palmer, 1995). Factors capturing human and financial capital, and in particular education,
have, indeed, been found to be important predictors of support for EU integration
(Hakhverdian et al., 2013).
At the collective level, a utilitarian cost-benefit analysis predicts that people living in
countries that gain from European integration in terms of net fiscal transfers, trade expansion,
more prominent stance in international relations, and so on will be more supportive of it
(Anderson and Reichert, 1995; Eichenberg and Dalton, 1993). The utilitarian motives also
find reflection in studies of EU sentiment in aspiring and new EU member states. Socio-
8
economic factors (income, education) matter in explaining support for the EU (e.g.,
Caplanova et al., 2004), and, in addition, ‘winners’ from post-communist transition (captured
by self-reported improvement of financial situation) and supporters of free-market reforms
tend to be more pro-European (Tucker et al., 2002; Caplanova et al., 2004; Loveless, 2010).
The main proposition of the identity approach is that attitudes toward EU integration are
driven by “group attachments, loyalties, values, and norms that define who a person is”
(Hooghe and Marks, 2004, p.415). European integration leads to a pooling of sovereignty that
potentially erodes national self-determination; therefore, people with stronger attachment to
their country, national identity, and national pride are expected have more negative attitudes
toward EU integration (Hobolt and De Vries, 2016). Empirically, identity-related factors have
been shown to be at least as important as utilitarian considerations in predicting support for
European integration (Carey, 2002; Hooghe and Marks, 2004), although the identity motives
have not been explicitly tested in studies on new and aspiring EU member states (Tucker et
al., 2002; Caplanova et al., 2004; Loveless, 2010). It should further be noted that different
identities are not exclusive – people may have local/regional, national, and European
identities at the same time – and that inclusive identities are more likely to be positively
linked to support for European integration relative to exclusive national identities (Hooghe
and Marks, 2005). It is also not clear to what extent European identity is endogenous to
support for the EU, as the latter may be causing the former (Hobolt and De Vries, 2016).
The basis of the third set of explanations related to cue-taking is that most people have little
direct personal experience with the EU and, therefore, limited knowledge about it (Anderson,
1998). In the absence of direct information, people rely on proxies, or informational
9
‘shortcuts,’ to evaluate the EU (Hobolt and De Vries, 2016). Such cues are often provided by
domestic politics: given that national governments typically negotiate integration processes
and people tend to follow national politics closely (Hobolt and De Vries, 2016) the
evaluations of the performance of national governments, leaders, and parties are projected
onto European integration (Franklin et al., 1995, Hobolt and De Vries, 2016). Empirical
studies indeed show that greater support for domestic governments is associated with greater
support for European integration (Anderson, 1998; Franklin et al., 1995).5 Apart from
domestic politics, people may also rely on social networks– relatives, friends, colleagues, and
neighbors – for informational shortcuts (Tanasoiu and Colonescu, 2008). Regardless of who
provides the cues, the question of information content is important as well. While it has been
argued that any information about the EU promotes support for integration (Inglehart et al.,
1991) because people who are more familiar with the process of European integration
become less threatened by it, it is reasonable to assume that negative information about the
EU will lead to less favorable attitudes about European integration. This assumption is
supported by empirical studies showing that the way in which national media frame European
integration is related to people’s attitudes toward the European project itself (Azrout et al.,
2012; Schuck and De Vreese, 2006; Vliegenthart et al., 2006).
2.2. Migration and support for joining the EU
5 However, an inverse relationship is also possible, as people dissatisfied with their own governments’
performance may be willing to delegate more decision-making power to the EU, and vice versa (Sanchez-
Cuenca, 2000).
10
This section outlines our expectations of how migration-related phenomena – willingness to
migrate, having relative abroad, and being a return migrant – might affect individual attitudes
toward joining the EU. We differentiate between migration to EU and to non-EU countries
and discuss each type in turn.
2.2.1. Willingness to migrate to an EU country
Being part of the EU can bring a number of tangible benefits, the most obvious being visa-
free travel to west European countries, many of which are primary destinations for
prospective migrants in the Western Balkans (Ivlevs and King, 2017). Visa-free travel would
be desirable for job seekers, for those working abroad and visiting their families back home,
and for family members residing in the Western Balkans and visiting relatives in the EU. In
addition, prospective migrants might expect that sooner or later, EU membership would bring
legal, work-permit free employment opportunities and access to social services. These
motivations are consistent with the utilitarian, or cost-benefit, approach, outlined earlier, for
explaining people’s support for European integration (Gabel and Palmer, 1995): certain
groups of people, in our case prospective migrants, are more likely to support European
integration because they expect personal benefits from it.
It is also likely that in the process of preparation to move to an EU country, prospective
migrants collect information about their destination and, as a result, become better informed
about the EU. As discussed in the previous section, given that information about the EU
increases support for the European project, probably because people more familiar with the
EU are less threatened by it (Inglehart, 1991; Janssen 1991), one can expect that willingness
11
to migrate also indirectly affects support for joining the EU through having better information
about the EU.
These considerations lend themselves to the following set of testable hypotheses:
H1a: People planning to emigrate to an EU member state are more likely to support
their own country’s EU accession.
H1b: People planning to emigrate to an EU member state are better informed about
the EU.
2.2.2. Relatives abroad
The EU preferences of people whose relatives migrated to the EU may be driven by several,
potentially conflicting, channels of influence. Separation from a family member typically
brings some unhappiness to those staying behind: for example, Graham et al. (2015), Ivlevs
et al. (2019), Marchetti-Mercer (2012) and Nobles et al. (2015) document that migrants’
family members staying in the origin countries report greater-than-average stress and
depression. Whether EU accession would intensify this feeling is unclear. If EU membership
allows migrants to find better jobs more easily, stay abroad longer, or bring other family
members with them, those left behind may feel negatively affected. However, if EU
membership brings more travel and visiting opportunities, those staying behind could feel
positively affected. It is also possible that migrants’ relatives gain utility from knowing that
migrants fulfil their potential abroad to a degree that could not be achieved in the home
country (Ivlevs et al. 2019). Receiving remittances is an additional factor that can affect
migrant relatives’ attitudes toward joining the EU. Remittances represent a non-negligible
income gain in the Western Balkan countries (Ivlevs and King, 2017), and EU membership
12
could increase remittances through better work opportunities for migrants and more reliable
and cheaper ways of transferring their remittances. Receiving remittances has been shown to
boost the evaluations of the best possible life for migrant household members staying behind,
above and beyond having relatives abroad (Ivlevs et al., 2019), meaning that people in
remittance-receiving households would support deeper EU integration.
On balance, if the positive effects associated with travelling, fulfilment, and remittances
outweigh the negative ‘separation’ effect, people with relatives residing in the EU would
favour EU membership. If, however, the separation effect is stronger, migrant relatives would
oppose further integration with the EU. Either way, migrant relatives’ support for joining the
EU would be driven by utilitarian cost-benefit considerations. Besides utilitarian motives,
migrant relatives may also become more pro-European indirectly because they are better
informed about the EU. Being better informed may come from migrants’ communication
with family members back home, from direct experience via visiting EU-resident migrants, or
a combination thereof. This argument links with the importance of information in the cue-
taking approach to explaining support for European integration: migration expands the range
of information sources about the EU, probably leading to more direct and varied knowledge
about the bloc.
These considerations lead us to the second set of hypotheses:
H2a: People whose relatives reside in an EU country are more likely to support their
country’s membership in the EU.
H2b: People whose relatives reside in an EU country are better informed about the EU.
13
2.2.3. Return migrants
Return migrants have had the opportunity to acquire substantial experiential information
about the EU member state(s) in which they resided. They may have experienced better
governance, institutions, rule of law, social protection, and more economic opportunities
compared to their home country. If the expectation is that these better quality institutions are
transferred to the home country via accession, the returning migrant may be more likely to
support EU membership. It is, however, also possible that the return migrant may have
experienced some of the EU’s darker sides, such as high inequality, high social segmentation,
high prices, and discrimination in wages and housing; the decision to return may indeed have
been driven by bad experiences. Such experiences, projected onto their home countries, may
account for a more reserved attitude toward EU membership.
On the instrumental side of the motivation spectrum, one particular reason why EU return
migrants might favor their country’s accession to the EU is related the transfer of pension
payments. Return migrants with employment history in one or several EU member states face
fragmented pension and social security entitlements (Holzmann et al., 2005) and may look
toward their home country’s EU accession as a way to simplify the collection of future
retirement benefits. Furthermore, return migrants may be considering another phase of work-
related migration to the EU. EU accession would make it simpler to keep this option open.
Finally, return migrants are particularly likely to be well-informed about the EU, which, as
already mentioned, by itself should increase their support for European integration.
These considerations lead us to the third set of hypotheses:
H3a: People with migration experience to an EU country are more likely to support their
country’s membership in the EU.
14
H3b: People with migration experience to an EU country are better informed about the EU.
2.2.4. Migration to the EU versus migration to non-EU countries
So far, we have looked at migration specifically to the EU. But will migration to non-EU
countries also dispose people favorably toward joining the EU? For example, people aspiring
to emigrate to or visit, say, Canada might find their home country’s EU membership to be a
helpful ‘jumping board’ for moving to Canada. Indeed, most ‘new’ EU members states
started enjoying visa-free travel to Australia, Canada, and the United States after joining the
EU in 2004,6 and some (e.g., the Baltic States, Slovakia) have been described as “winners” in
terms of global mobility rights (Mau et al., 2015). The willingness to join the EU in order to
be able to visit or migrate elsewhere follows the ‘utilitarian’ logic in explaining attitudes
toward European integration (Gabel and Palmer, 1995; Hobolt and De Vries, 2016). It is also
possible that people considering emigration or associated with the diaspora are favorably
inclined to any developments that increase transnational interaction, openness, and choices
and, therefore, favorably inclined toward the EU. This situation would reflect the
cosmopolitan outlook associated with migration (Brown, 2018; Cook, 2012; Nedelcu, 2012).
We formalize these considerations in the following hypothesis:
6 For example, Canada lifted travel restrictions for Estonia in 2006, Latvia and the Czech Republic in 2007, and
Poland, Slovakia, Lithuania, and Hungary in 2008; Canada also has easier access to work permits for visa-
exempt citizens. Australia lifted travel restrictions for new EU member states in 2008, as did the United States
in 2009 (except Poland). Source: DEMIG POLICY database (https://www.imi.ox.ac.uk/data/demig-data/demig-
policy-1) and the European Visa Database (http://www.mogenshobolth.dk/evd/default.aspx) .
15
H4a: People willing to emigrate to a non-EU country are more likely to support their
country’s membership in the EU.
H4b: People with relatives residing in a non-EU country are more likely to support their
country’s membership in the EU.
H4c: Migrants who have returned from a non-EU country are more likely to support their
country’s membership in the EU.
3. DATA, VARIABLES, AND EMPIRICAL MODELS TO BE TESTED
3.1. Data
To test our hypotheses, we use data from the Gallup Balkan Monitor survey. The survey, best
described as repeated cross-sections, was conducted in Albania, Bosnia and Herzegovina,
Croatia, Kosovo, Macedonia (FYROM), Montenegro, and Serbia in 2006, 2008, 2009, 2010,
and 2011 by the international opinion poll agency, Gallup. The core Gallup World Poll
questionnaire was augmented with questions relevant to the Western Balkan countries and
translated into the respective countries’ languages. Face-to-face interviews, taking
approximately 45 minutes, were conducted at respondents’ homes and restricted to one member
per household. Approximately 1,000 people were interviewed in each country each year.
The samples of respondents were designed to be nationally representative and followed a three-
stage probability-based respondent selection methodology. In the first stage, Primary Sampling
Units (PSUs) were randomly selected from a pool of PSUs based on census, administrative,
and migration information. In the second stage, households were selected with the help of a
16
standard random route technique. In the third stage, respondents within households were
selected using either Kish-grid or the most recent birthday rule.7 Certain politically sensitive
regions (e.g., Northern Serb enclaves in Kosovo, Republic Srpska in Bosnia and Herzegovina)
were intentionally overrepresented in the sample; this is why in our analysis we apply the
weighting proposed by Gallup for the purpose of re-balancing. More information on survey
methodology and implementation is available in Gallup (2010).
Given that some key questions, such as a question on the experience of living abroad and a
question on voting in a hypothetical EU referendum, were asked only in 2009, 2010, and 2011,
we exclude years 2006 and 2008 from our analysis.
3.2. Variables
3.2.1 Dependent variable(s): support for joining the EU
We use two questions to capture public support for joining the EU. First, respondents were
asked: “If there was a referendum on Sunday on joining the EU, what would you vote?”
Possible answers included “[Country] should join” and “[Country] should not join.” We
created a binary variable vote yes in an EU referendum, taking the value ‘1’ if the
respondent thought the country should join and ‘0’ if the respondent thought the country
should not join.8 Overall, for the years 2009-11, 84% of respondents would have voted ‘yes’
7 Under the most recent birthday rule, the survey interview is conducted with a household member who had the
most recent birthday. This approach assures a random selection of respondents from households.
8 We excluded respondents who answered “don’t know” (9.8%) or refused to answer (1.4%). It is, however,
possible that excluding a significant proportion of respondents might result in a selection bias, which is why we
17
(and 16% ‘no’). At the country level, the share of positive vote was 50% in Croatia, 74% in
Serbia, 87% in Bosnia and Herzegovina, 88% in FYR Macedonia, 91% in Montenegro, 95%
in Kosovo, and 97% in Albania.
Second, respondents were asked: “Do you think [Country’s] membership in the EU would be
a good thing?,” with possible answers “a bad thing,” “neither good nor bad thing,” and “a
good thing.” We created a categorical variable EU membership good, taking the value ‘1’ if
the answer was “bad,” ‘2’ if “neither good nor bad,” and ‘3’ if “good.”9 The whole-sample
distribution of answers is 11%, 23%, and 66%, respectively. The share of those who
considered EU membership a good thing was 29% in Croatia, 46% in Serbia, 64% in FYR
Macedonia, 66% in Bosnia and Herzegovina, 74% in Montenegro, 89% in Albania, and 91%
in Kosovo.
Although one would expect a high correlation between the two variables (in our case, the
Pearson correlation coefficient is 0.74, and the polychoric correlation coefficient is 0.92),
there may be people who would vote ‘yes’ in a referendum but not necessarily consider their
country’s membership in the EU a good thing. Arguably, a vote in a hypothetical referendum
could draw on personal and instrumental considerations to a greater extent than the answer to
estimated additional models where the dependent variable incorporated excluded respondents. In particular, we
merged the “don’t know” and refused to answer categories and added them as a ‘neutral’ category in an ordered
probit model and as a third category in a multinomial probit model. The results, available on request, are
consistent with the model which included only ‘yes’ and ‘no’ answers.
9 We again exclude the ‘don’t know’ answers (5.1%) and refusals (0.4%). In a robustness check, we merged
them with the ‘neither good nor bad’ category and obtained qualitatively similar results (available on request).
18
the question on whether the EU membership was a good thing. The latter could be interpreted
more broadly and include considerations of EU membership’s effects on shaping the
country’s future.
3.2.2. Dependent variable: Being well informed about the EU
The variable being well informed about the EU is based on the question, “Please rate how
much you are informed about the European Union.” In accordance with possible answers to
this question, the variable takes four values from 1 (not at all informed) to 4 (very well
informed).10
3.2.3 Main regressors: migration-related variables
We created three migration-related variables capturing 1) willing to migrate to an EU country
in the future, 2) having relatives in an EU country, and 3) having previous experience of
living in an EU country. The first variable, willing to move to the EU, is based on the
question, “Ideally, if you had the opportunity, would you like to move permanently to another
country, or would you prefer to continue living in this country?” If respondents said they
wanted to move, they were subsequently prompted to indicate a preferred destination. If the
indicated country belonged to the EU-27, the answer was coded as 1, in all other cases (prefer
10 The summary statistics for this and other variables are available in Table A1 of the Supplementary Appendix.
19
to continue living in this country/would like to move permanently to a country other than the
EU-27 member states) the answer was coded as 0.11
The variable relatives in the EU draws on the question, “Do you have relatives or friends
who are living in another country whom you can count on to help you when you need them,
or not?” If the respondent said yes, s/he was asked to indicate in which country such relatives
or friends lived. The variable was coded 1 if the indicated country was an EU-27 member
state and 0 in all other cases.
The variable been in the EU draws on the question, “Have you already studied or worked
abroad?” For positive answers, the country of work or study was solicited. The variable took
value 1 if the indicated country belonged to the EU-27 and 0 otherwise (had not worked or
studied abroad or had worked or studied in a country other than an EU member state).
Finally, to test hypotheses H4a-H4c, we created three dummy variables capturing migration
to a non-EU country: 1) willing to move to a non-EU country, 2) having relatives/friends in
a non-EU country, and 3) been in a non-EU country for work or study.
3.2.4. Control variables
11 While a natural criticism of such emigration willingness/intentions variables is that intentions do not
necessarily follow actual moves, longitudinal studies suggest that the two are highly correlated. See Ivlevs
(2015) for the summary of the evidence and limitations of emigration willingness/intentions variables.
20
Following the literature on individual-level determinants of attitudes toward European
integration (e.g., Caplanova et al., 2004; Loveless, 2010; Tucker et al., 2002; Tanasoiu and
Colonescu, 2008; Jackson, 2011), all regressions included the following standard socio-
demographic controls: age and age squared, gender, education levels (primary, secondary,
and tertiary), income levels (dummies for within-country household per capita income
quintiles and a dummy for non-reported income), subjectively evaluated change in the
standard of living (better, the same, or worse), being unemployed, and religious
denominations (Catholic, Orthodox, Muslim, and other).
In addition, we controlled for support for national government and strong European
identity, both of which have been shown to be significant predictors of public support for
European integration (Franklin et al., 1995; Ray, 2003; Tanasoiu and Colonescu, 2008;
McLaren, 2004; Hooghe and Marks, 2004). These variables are based on the questions: a)
“Please indicate how good of a job the [Country’s] government does?,” with possible answers
‘poor’ (1), ‘fair’ (2), ‘good’ (3), and ‘excellent’ (4), and b) “How strongly do you identify
yourself with Europe?,” with possible answers ranging from ‘not at all’ (1) to ‘extremely
strongly’ (5).
Focusing on the Western Balkan context, we also wanted to control for individual
assessments of the international community’s role in the region’s evolving politics. For
example, people in the Western Balkans may have diverging opinions about the international
community’s involvement in post-Yugoslavian military conflicts and the process of Kosovo’s
independence. As EU countries played a dominant role in the region’s politics, people’s
assessments of that role might have influenced their attitudes toward joining the EU. We,
21
thus, created a variable, helpful international community, which draws on the question:
“Thinking of the last 15-20 years and taking everything into account, overall how would you
describe the International community’s role [in your country]?” The variable took values 1
(extremely harmful) to 5 (extremely helpful).
Finally, to control for all year- and country-specific influences on both the migration
variables and attitudes toward joining the EU, we included year- and country-fixed effects
(dichotomous variables for each country and each year).12 Among other things, the inclusion
of country-fixed effects ensures that the relationship between the variables of interest is
estimated at the within-country level and not driven by differences between countries.
3.3. Empirical models to be tested
We model support for EU membership of individual i living in country j as follows:
Support for EU membershipi.j = α0 + α1* migration-related variablesi,j +
α2* informed about the EUi,j +
α3* controlsi,j +
α4* year fixed effects+
α5* country fixed effectsj +
12 As a robustness check, we also included country-year fixed effects (a dichotomous variable for each country-
year pair), which would account for the effect of temporal changes in country-level variables on attitudes toward
joining the EU. Our results regarding the link between emigration and attitudes toward joining the EU remained
essentially unchanged.
22
idiosyncratic error termi,j (1)
where migration-related variables relate to migration to the EU and α0 - α5 are the parameters
(parameter vectors) to be estimated.
Note that the migration-related variables and the variable informed about the EU are all
included as regressors in Model 1. Such model specification allows us to separate migration-
related variables' utilitarian effect from ‘informedness’ about the EU, which might also be
driven by migration. Migration-related variables’ influence on ‘informedness’ about the EU
is tested, using the following model:
Well informed about the EUi.j = α0 + α1* migration-related variablesi,j +
α2* controlsi,j +
α3* year fixed effects +
α4* country fixed effectsj +
idiosyncratic error termi,j (2)
Finally, Model 3 focuses on the effects of variables capturing migration to a non-EU country
on support for EU membership. Among other things, this model tests for a ‘cosmopolitan
effect’: the possibility that international mobility and orientation in general make people more
positively inclined toward joining international collaborations, such as the EU. Note that
variables capturing migration to the EU and ‘informedness’ about the EU are also included,
leaving people with no connection to migration as a reference group.
23
Support for EU membershipi.j = α0 + α1* migration-related variables (non-EU)i,j +
α2* migration-related variables (EU)i,j +
α3* informed about the EUi,j +
α4* controlsi,j +
α5* year fixed effects +
α6* country fixed effectsj +
idiosyncratic error termi,j (3)
In the case of a binary dependent variable (vote yes in an EU referendum), we estimate the
model with binary probit. Where the dependent variable is categorical and ordered (EU
membership a good thing and well informed about the EU), we use ordered probit.13
Given a potential correlation within the group of migration-related variables (e.g., people
may be more likely to express willingness to migrate if they have past migration experience
or relatives abroad), we first include the three migration-related variables separately and then
estimate a model where they are included jointly.
4. RESULTS
Table 1 presents the correlates of attitudes toward EU membership. To save space, here and
in what follows, we present only the coefficient of the variables that are central to our
13 As a robustness check, we estimated our models with logit and OLS, and the results are consistent with the
probit estimations.
24
analysis (migration-related variables and ‘informedness’ about the EU); the coefficients of
control variables are available in the Supplementary Appendix (Tables A2-A4). All three
migration-related variables are positive and statistically significant predictors of the
likelihood of voting ‘yes’ in a hypothetical EU referendum (Panel A). Considering the case
where migration variables are included separately (Columns 1-3) and keeping other factors
constant, people willing to migrate to the EU are 2.5 percentage points more likely, those
with relatives in the EU 2.1 percentage points more likely, and EU return migrants 3.1
percentage points more likely to say they would vote favorably in an EU referendum. The
magnitudes of these marginal effects slightly decline (to 2.1, 1.8, and 2.5 percentage points,
respectively) when the three migration variables are included jointly (Column 4). This
reflects a positive correlation between them, which is to be expected. However, even when
included jointly, the three variables remain positive and statistically significant. This finding
indicates that each migration-related variable exerts a positive influence on the likelihood of
voting ‘yes,’ regardless of the other two migration-related variables' effect.
Panel B of Table 1 reports the results for the model explaining the likelihood of considering
membership of one’s country in the EU a good thing. Regardless of whether included jointly
or separately, the variables capturing willingness to emigrate to the EU and return migration
are statistically insignificant. Only people with relatives or friends in the EU are more likely
to say that their country’s membership in the EU would be a good thing. In terms of marginal
effects, relatives of EU migrants are 1.1 percentage point less likely to consider EU
membership a bad thing, 0.9 percentage point less likely to consider EU membership neither
good nor bad, and 1.9 percentage points more likely to consider EU membership a good
25
thing.14 These results provide mixed support for H1a, H2a, and H3a. On the one hand,
respondents from all three migration-related categories are more likely to vote favorably in a
hypothetical EU referendum. On the other hand, only EU migrants’ relatives are more likely
to consider that their country’s membership in the EU would be a good thing, while
prospective and return EU migrants do not distinguish themselves from the rest of the
population.
Table 1: Migration and support for country’s membership in the EU
A. Dependent variable:
Voting yes in a hypothetical EU referendum; binary probit marginal effects
B. Dependent variable: EU membership a good thing;
ordered probit coefficients (1) (2) (3) (4) (5) (6) (7) (8)
Willing to move to the EU 0.025*** - - 0.021** 0.067 - - 0.051 Relatives in the EU - 0.021*** - 0.018*** - 0.079** - 0.070** EU return migrant - - 0.031*** 0.025** - - 0.097 0.074 Well informed about the EU 0.013*** 0.013*** 0.013*** 0.013*** 0.210*** 0.209*** 0.208*** 0.207*** Individual-level controls Yes Yes Yes Yes Yes Yes Yes Yes Country-fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Year-fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Observations 16,858 16,858 16,858 16,858 17,871 17,871 17,871 17,871 Pseudo R2 0.257 0.257 0.257 0.258 0.191 0.191 0.191 0.191 Chi2 2084 2085 2082 2125 2761 2761 2761 2762
Notes: *** p<0.01, ** p<0.05, * p<0.1, heteroscedascity-robust standard errors used to calculate regressors’ level of significance. See Table A2 of the Supplementary Appendix for complete econometric output. Source: Gallup Balkan Monitor survey and authors’ calculations.
Note that the effects of the migration-related variables, presented above, have been obtained
while keeping ‘informedness about the EU’ constant. ‘Informedness’ was included as a
14 These marginal effects are based on the estimated coefficient of the relatives in the EU variable in
specification 8.
26
control in all regressions of Table 1, and, as expected, being well informed about the EU is a
positive and statistically significant predictor of positive attitudes toward EU membership
(both voting favorably in an EU referendum and considering EU membership a good thing).
We have, however, also hypothesized that migration-related variables can have an
independent effect on being well informed about the EU. The empirical test of this hypothesis
is presented in Table 2. Having worked or studied in the EU has a positive and statistically
significant correlation with the probability of being well informed about the EU. The
coefficient of the relatives-in-the-EU variable is positive but statistically insignificant.
Interestingly, controlling for work/study experience in the EU and for having relatives in the
EU, people willing to move to the EU are somewhat less likely to say they are well informed
about the EU (Column 4 of Table 2). The results thus provide support for H3b, but not H1b
or H2b.
27
Table 2: Migration and informedness about the EU
Dependent variable: Well informed about the EU;
ordered probit coefficients (1) (2) (3) (4)
Willing to move to the EU -0.031 - - -0.056* Relatives in the EU - 0.029 - 0.019 EU return migrant - - 0.226*** 0.233*** Individual-level controls Yes Yes Yes Yes Country-fixed effects Yes Yes Yes Yes Year-fixed effects Yes Yes Yes Yes Observations 18,539 18,539 18,539 18,539 Pseudo R2 0.0550 0.0551 0.0560 0.0562 Chi2 1363 1366 1396 1398
Notes: *** p<0.01, ** p<0.05, * p<0.1, heteroscedascity-robust standard errors used to calculate regressors’ level of significance. See Table A3 of the Supplementary Appendix for complete econometric output. Source: Gallup Balkan Monitor survey and authors’ calculations.
With a final set of regressions, we test the hypotheses that variables capturing migration to
non-EU countries are positively associated with support for EU accession (Model 3, H4a-
H4c). The results are presented in Table 3. People considering emigration to a non-EU
country are 2.2 percentage points more likely to vote favorably in a hypothetical EU
referendum (Columns 1 and 4). They are also more likely to consider EU membership a good
thing (columns 5 and 8); in terms of marginal effects, prospective migrants to non-EU
countries are 2 and 1.7 percentage points less likely consider EU membership a ‘bad’ and,
respectively, ‘neither good nor bad’ thing, and 3.7 percentage points15 more likely to consider
EU membership a ‘good’ thing. These results strongly support H4a that people considering
emigration to a non-EU destination are more likely to support their country’s accession to the
EU.
15 These marginal effects were calculated after estimating specification 8 of Table 3.
28
The estimates of the two other variables of interest (having relatives in a non-EU country and
having worked/studied in a non-EU country) are statistically insignificant, regardless of the
way attitudes toward EU accession are captured (voting in a referendum or considering EU
membership a good thing) and whether the regressors of interest are included in the model
jointly or separately. H4b and H4c are, thus, not supported by the data.
Finally, note that the results of the migration-to-the-EU variables, which were also included
in Model 3 and reported in Table 3, remain qualitatively unchanged.
Table 3: Migration to non-EU destinations and attitudes towards EU membership
A. Dependent variable:
Voting yes in a hypothetical EU referendum; binary probit marginal effects
B. Dependent variable: EU membership a good thing;
ordered probit coefficients (1) (2) (3) (4) (5) (6) (7) (8) Willing to move (non-EU) 0.022** - - 0.022** 0.129*** - - 0.136*** Relatives abroad (non-EU) - 0.007 - 0.006 - -0.050 - -0.056 Return migrant (non-EU) - - -0.017 -0.021 - - -0.069 -0.077 Willing to move (EU) 0.027*** - - 0.023** 0.082* - - 0.066 Relatives abroad (EU) - 0.022*** - 0.018*** - 0.072** - 0.061* Return migrant (EU) - - 0.031*** 0.024** - - 0.094 0.071 Individual-level controls Yes Yes Yes Yes Yes Yes Yes Yes Country-fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Year-fixed effects Yes Yes Yes Yes Yes Yes Yes Yes Observations 16,858 16,858 16,858 16,858 17,871 17,871 17,871 17,871 Pseudo R2 0.257 0.257 0.257 0.259 0.191 0.191 0.191 0.191 Chi2 2093 2088 2080 2135 2778 2760 2759 2776
Notes: *** p<0.01, ** p<0.05, * p<0.1, heteroscedascity-robust standard errors used to calculate regressors’ level of significance. The set of control variables includes informedness about the EU. See Table A4 of the Supplementary Appendix for complete econometric output. Source: Gallup Balkan Monitor survey and authors’ calculations.
5. DISCUSSION AND CONCLUSION
29
This study set out to assess whether emigration affects attitudes toward joining the EU in
prospective EU member states - Albania and the successor states of Yugoslavia. To answer
our research question, we used data from the Gallup Balkan Monitor survey, a large and
underexplored dataset containing rich information on attitudes toward joining the EU and
migration in the Western Balkan countries. We first discussed the ways in which migration
can affect attitudes toward EU membership and then estimated the relationship between
migration-related variables – willingness to migrate, having relatives abroad, and being a
return migrant – and the likelihood of providing an affirmative answer to the following two
questions:
(i) “If there were a referendum on Sunday on joining the EU, what would you vote?”
(ii) “Do you think [your country’s] membership in the EU would be a good thing?”
The results revealed intriguing differences in migration’s effects on responses to these, on
first sight, similar questions regarding public support for the EU. Prospective migrants, return
migrants, and people with a migrant family member or close friend in the EU are, across the
board, more likely to vote yes in a hypothetical referendum on whether to join the EU.
Responses to the second question, however, deliver a less clear picture. Only people who had
relatives or close friends in the EU were more likely to think that their country’s membership
in the EU would be a good thing. Return migrants and prospective migrants were, in this
regard, indistinguishable from people without migrant connections.
One possible interpretation of this situation is that a vote in a national referendum would
draw relatively more strongly on personal utilitarian/instrumental motives, related to
30
anticipated opportunities for easier job search and travel. The question, whether one’s
country’s membership in the EU would be a good thing, lends itself to broader interpretation,
including one’s view on whether membership would be a good thing for the country’s future
development. People may doubt that their country would have a good future inside the EU
but still vote ‘yes’ in a hypothetical referendum because they expect concrete benefits for
themselves. Also, for those willing to emigrate permanently, their home country’s future may
be of lesser concern. Overall, these results suggest that utilitarian considerations play a
significant role in explaining public support for joining the EU in the Western Balkans, which
in turn advances theory by adding migration-related variables to the battery of factors
(income, education etc.) through which utilitarian considerations can manifest themselves
(Gabel and Palmer, 1995; Hobolt and De Vries, 2016). Also, differential responses to the
‘twin questions’ helps sharpen the split between self-centred instrumental motives and
sociotropic concerns.
We also tested how migration-related variables correspond to being well informed about the
EU, which by itself can positively affect pro-EU sentiment. Not surprisingly, a strong
positive information effect is seen in people who lived or studied in the EU. Having a migrant
relative or close friend in the EU, however, does not show any significant influence on
‘informedness,’ suggesting that information transfer through relatives or friends residing in
the EU is of limited effectiveness. In terms of the cue-taking approach to explaining public
support for the EU (Hobolt and De Vries, 2016), our results show that relatives and friends
residing in the EU may not be a source of informational ‘shortcuts’ about the EU (see
Tanasoiu and Colonescu, 2008). Similar lack of significant effects is observed for
prospective EU migrants. Here, an explanation may be that people with concrete plans to
31
emigrate face concrete questions, which make them painfully aware of all the things they do
not know. They do not feel well informed about the EU because they apply more
discriminating standards of ‘informedness.’ Overall, we obtain limited support for
migration’s potential ‘informational effect’ on joining the EU: it works for return migrants
but not for prospective migrants or migrant relatives and friends staying in home countries.
To test for migration’s potential ‘cosmopolitan effect’ on joining the EU, we looked at
attitudes toward EU membership of people with migration connections to countries outside
the EU. Support for the ‘cosmopolitan’ effect is limited. People who lived or studied in a
non-EU country, as well as those whose relatives resided in a non-EU country, were not
found to be more likely to vote favorably in a hypothetical EU referendum or to consider EU
membership a good thing. Only people willing to emigrate to a non-EU country show
evidence of higher support for their country’s EU membership. This finding, however, could
indicate instrumental motives at work, insofar as EU membership may be seen as a ‘jumping
board’ for emigration to countries other than EU.
Where do these findings leave us? This article contributes to scholarly discourse in several
ways. First, the existing literature on migration’s influence on support for European
integration has focused on perspectives from ‘old’ EU member states, within-EU mobility,
and migration from outside Europe (Azrout et al., 2011; De Vreese et al. 2008; Hobolt et al.
2011; Gerhards and Hans, 2011; McLaren, 2007; Kuhn, 2015), leaving the perspectives of
countries aspiring to join the EU unexplored. We have filled this gap by focusing on the links
between emigration and public opinion on European integration in the Western Balkans –
countries next in line to join the EU.
32
Second, our study adds to the broader literature on the determinants of EU preferences in
prospective member states. There is substantial evidence on the individual-level determinants
of attitudes toward EU membership in countries that joined the EU in 2004 and 2007 (e.g.,
Caplanova et al., 2004; Loveless, 2010; Tucker et al., 2002; Tanasoiu and Colonescu, 2008;
Jackson, 2011). These contributions have sought to establish how support for joining the EU
is affected by factors particularly relevant in the transition context, such as attitudes to free
markets, experience of transition from planned to market economy, and perceptions of social
inequality. Despite the salience of emigration from the Western Balkan region, variables
capturing migration-related phenomena have never entered the analysis. Our study, thus,
offers an explicit treatment of emigration-related variables – willingness to emigrate, having
relatives abroad, and being a return migrant – as potential antecedents of people’s support for
joining the EU.
Finally, our study contributes to the literature on international migration’s effects on the
development of political institutions in migrants’ origin countries. Closely related to the
concept of ‘social remittances’ (Levitt, 1998),16 this literature has shown that emigration can
enhance government accountability, political participation, and voting for pro-democratic
parties and reduce public sector corruption in migrants’ home countries (Barsbai et al., 2018;
Batista and Vicente, 2011; Chauvet and Mercier, 2014; Docquier et al., 2016; Ivlevs and
King, 2017; Pfutze, 2012). Our study adds to this literature by showing that prospective and
return migrants, as well as migrant relatives staying in the home countries, are more likely to
support joining a supranational institution, such as the EU. Joining the EU, in turn, is likely to
foster political and institutional development in these countries. Specifically, as a condition
16 Levitt (1998) defines social remittances as “ideas, practices, identities, and social capital that flow from receiving- to sending-country communities” (p.926).
33
for joining the EU, candidate countries must ensure the existence of institutions that
guarantee democracy, the rule of law, human rights, and protection of national minorities
(Pridham, 2005; Schimmelfennig and Sedelmeier, 2008). Evidence suggests that the pressure
of such political conditionality indeed contributed to democratization and institutional
reforms in Central and Eastern European countries that joined the EU in 2004 (Elbasani,
2013; Pop-Eleches, 2007; Pridham, 2005; Schimmelfennig, 2007). Therefore, through raising
support for joining the EU, emigration is likely to contribute to the political and institutional
development of the successor states of Yugoslavia.
While we uncover an important relationship between migration, especially to and from the
EU, and favourable attitudes toward joining the bloc, there are several limitations to our
study. First, our results should be interpreted as conditional correlations rather than causal
effects. Even after controlling for a wide range of factors, there may remain unobserved
individual and/or household-level characteristics driving both migration and attitudes toward
the EU. One way to move closer to causality, and a possible direction for future research,
would be to use panel data, where migration histories and EU preferences of the same people
are observed over time. Second, a question on whether EU membership is a good thing,
which we use to capture favorable attitudes toward joining the EU, is very broad. While it
may serve the purpose of distilling a net sentiment from a multitude of more narrowly
targeted and possibly contradicting sentiments, responses to such a broad question must be
treated with caution. A solution here could be to include in surveys a range of ‘good thing for
whom’ questions, asking whether EU membership would be good for respondents
themselves, their family members, or their country as a whole. Third, we have hypothesized
that migration can affect pro-EU sentiment indirectly through being better informed about the
34
EU. A similar argument could go for European identity: for example, return EU migrants or
people whose relatives reside in the EU may develop a stronger European identity, which
would feed into a favorable stance toward joining the EU.17 Assessing migration’s
importance for European identity in prospective EU member states is beyond this study’s
scope and left for future research.
All things considered, our results show that migration is good for fostering positive attitudes
toward joining the EU among the prospective member states of the Western Balkans. In the
presence of seemingly pervasive Euroscepticism, migration can play a role in keeping the EU
desirable. Our results point toward a preponderance of instrumental motives for joining the
EU, compared to the ‘grand ideas’ of the EU’s formative stage. Recalling that the EU’s
creation was first and foremost about peace in Europe, the motives for joining seem to have
become ‘smaller.’ It may be an indicator of EU success, however, that one can afford to focus
on ‘small’ things, especially in the Western Balkans, where grander things like peace and
security may still not be taken for granted. Finally, in the context of a broader literature on
emigration’s effects on the development of migrant-sending countries, our study shows that
migration increases support for joining a supranational institution (the EU), which, in turn, is
likely to improve institutional quality in migrants’ origin countries.
17 See Verhaegen et al. (2014) on how perceived economic benefits may affect European identity.
35
REFERENCES
Anderson, C. J. (1998) “When in Doubt Use Proxies: Attitudes toward Domestic Politics and
Support for European Integration”, Comparative Political Studies 31(5): 569–601.
Anderson, C.J. and Reichert, M.S. (1995) "Economic Benefits and Support for Membership
in the E.U.: A Cross-National Analysis", Journal of Public Policy 15(3): 231-249.
Azrout, R., van Spanje, J., & de Vreese, C. (2011) “Talking Turkey: anti-immigrant attitudes
and their effect on support for Turkish membership of the EU”, European Union
Politics 12(1): 3-19
Azrout, R., van Spanje, J., & de Vreese, C. (2012). “When News Matters: Media Effects on
Public Support for European Union Enlargement in 21 Countries”, Journal of
Common Market Studies 50(5): 691–708.
Barsbai, T., Rapoport, H., Steinmayr, A. and Trebesch, C. (2017). The effect of labor
migration on the diffusion of democracy: Evidence from a former soviet republic.
American Economic Journal: Applied Economics 9(3): 36-69
Batista, C., and Vicente, P. (2011). “Do migrants improve governance at home? Evidence
from a voting experiment.” World Bank Economic Review 25(1): 77–104.
BBC (2012), “Croatian EU referendum: Voters back membership”, BBC News Europe,
22.01.2012. http://www.bbc.co.uk/news/world-europe-16670298 (accessed
22.07.2012)
36
Belloni, R. (2016) "The European Union Blowback? Euroscepticism and its Consequences in
the Western Balkans", Journal of Intervention and Statebuilding 10(4): 530-547.
Belloni, R., and Brunazzo, M. (2017) “After ‘Brexit:’ The Western Balkans in the European
Waiting Room”, European Review of International Studies 1: 21-38.
Brown, B. (2018) "Unlikely Cosmopolitans: An Ethnographic Reflection on Migration and
Belonging in Sri Lanka", Anthropological Quarterly 91(1): 209-236.
Brunnbauer, U. (2009) Labour Emigration from the Yugoslav Region from the late 19th
Century until the End of Socialism: Continuities and Changes. In Brunnbauer, U.
(ed.): Transnational Societies, Transterritorial Politics. Migrations in the (Post-
)Yugoslav Region, 19th–21st Century. Munich: R. Oldenbourg Verlag.
Caplanova, A., Orviska, M. and Hudson, J. (2008) “Eastern European Attitudes to Integration
with Western Europe.” Journal of Common Market Studies 42(2): 271-87.
Carey, S. (2002) "Undivided Loyalties: Is National Identity an Obstacle to European
Integration?", European Union Politics 3(4): 387-413.
Chauvet, L., and Mercier, M. (2014). “Do return migrants transfer political norms to their
origin country? Evidence from Mali.” Journal of Comparative Economics 42(3): 630–
657.
Cook, N. (2012) "Canadian development workers, transnational encounters and cultures of
cosmopolitanism", International Sociology 27(1): 3-20.
37
De Vreese, C. H., Boomgaarden, H. G. and Semetko, H. (2008) “Hard and soft: Public
support for Turkish membership in the EU”, European Union Politics 9 (4): 537-56.
Docquier, F., Lodigiani, E., Rapoport, H. and Schiff, M. (2016). “Emigration and
democracy”. Journal of Development Economics 120: 209–223.
Elbasani, A. (2013) European Integration and Transformation in the Western Balkans.
Europeanization or Business as Usual? Abingdon, Routledge.
Eichenberg, R.C. and Dalton, R.J. (1993) "Europeans and the European Community: the
dynamics of public support for European integration", International Organization
47(4): 507-534.
Favell, A. (2011) “Immigration, migration and free movement in the making of Europe”. In
Checkel, J. and Katzenstein, P. (eds.) European Identity. Cambridge University Press,
pp.167-189.
Franklin, M., van der Eijk, C. and Marsh, M. (1995) “Referendum Outcomes and Trust in
Government: Public Support for Europe in the Wake of Maastricht”, West European
Politics, 18(3): 101–117.
Gabel, M. J. (1998) “Public Support for European Integration: An Empirical Test of Five
Theories”, Journal of Politics 60(2): 333-354.
Gabel, M.J. and Palmer, H. (1995) "Understanding Variation in Public Support for European
Integration." European Journal of Political Research 27: 3-19.
38
Gallup (2010) ”Gallup Balkan Monitor, Insight and Perceptions: Voices of the Balkans.
2010, Summary of the Findings”. Available at:
https://www.esiweb.org/enlargement/wp-
content/uploads/2009/02/2010_Summary_of_Findings.pdf
Geddes A (2014) The EU, UKIP and the Politics of Immigration in Britain. The Political
Quarterly 85(3): 289–295.
Gerhards, J. and Hans, S. (2011) “Why not Turkey? Attitudes towards Turkish Membership
in the EU among Citizens in 27 European Countries”. Journal of Common Market
Studies 49(4): 741‐766.
Graham, E., Jordan, L.P. and Yeoh, B.S. (2015) “Parental migration and the mental health of
those who stay behind to care for children in South-East Asia.” Social Science and
Medicine 132: 225–235.
Hakhverdian, A., van Elsas, E., van der Brug, W. & Kuhn, T. (2013) "Euroscepticism and
education: a longitudinal study of 12 EU member states, 1973-2010", European
Union Politics 14(4): 522-541.
Halla, M., Wagner, A.F. and Zweimüller, J. (2017) "Immigration and Voting for the Far
Right", Journal of the European Economic Association, 15(6): 1341-1385.
Hobolt, S.B. and De Vries, C.E. (2016) "Public Support for European Integration", Annual
Review of Political Science 19(1): 413-432.
Holzmann, R., Koettl, J. and Chernetsky, T. (2005) “Portability Regimes of Pension and
Health Care Benefits for International Migrants: An Analysis of Issues and Good
39
Practices”, Social Protection discussion paper series No. 0519. Washington, DC:
World Bank.
Hooghe, L and Marks, G. (2004) “Does Identity or Economic Rationality Drive Public
Opinion on European Integration”, Political Science and Politics 37(3): 415-420.
Inglehart, Ronald (1970) “Cognitive Mobilization and European Identity”, Comparative
Politics 3: 45–70.
Inglehart, R, Rabier, J.-R, and Reif, K. (1991). "The Evolution of Public Attitudes toward
European Integration: 1970-86." In Eurobarometer: The Dynamics of European
Public Opinion, ed. K. Reif and R. Inglehart. London: Macmillan
Ivlevs, A. (2015) “Happy moves? Assessing the link between life satisfaction and emigration
intentions”, Kyklos 68(3): 335-356.
Ivlevs, A. and King, R. (2017) "Does emigration reduce corruption?", Public Choice 171 (3-
4): 389–408.
Ivlevs, A., Nikolova, M., and Graham, C. (2019) “Emigration, remittances, and subjective
well-being of those staying behind”, Journal of Population Economics 32(1): 113-
151.
Ivlevs, A. and Veliziotis, M. (2018). "Local-level immigration and life satisfaction: The EU
enlargement experience in England and Wales", Environment and Planning A 50(1):
175-193
40
Jackson, J. E., Bogdan W. M., and Miller-Gonzalez, J. (2011) “Buying Support and Regime
Change: The Evolution of Polish Attitudes towards the EU and Voting between
Accession and 2008,” European Union Politics 12 (2): 147-167.
Janssen, J. (1991). “Postmaterialism, cognitive mobilization and support for European
integration”. British Journal of Political Science 21: 443-457.
Ker-Lindsay, J. (2017) "The United Kingdom and EU enlargement in the Western Balkans:
from ardent champion of expansion to Post-Brexit irrelevance", Southeast European
and Black Sea Studies 17(4): 555-569.
Kuhn, T. (2015) Experiencing European Integration. Transnational Lives and European
Identity. Oxford University Press.
Lawless, J. (2015) “In an English market town, a political battle over immigration.”
Available at: http://www.macleans.ca/politics/worldpolitics/in-an-english-market-
town-a-political-battle-over-immigration/
Levitt, P. (1998). “Social remittances: Migration driven local-level forms of cultural
diffusion.” International Migration Review 32(4): 926–948.
Loveless, M (2010) “Agreeing in Principle: Utilitarianism and Economic Values as Support
for the European Union in Central and Eastern Europe”, Journal of Common Market
Studies 48(4): 1083–1106.
Marchetti-Mercer, M.C. (2012) “Those easily forgotten: the impact of emigration on those
left behind.” Family Process 51: 376–390.
41
Mau, S., Gülzau, F., Laube, L. and Zaun, N. (2015) "The Global Mobility Divide: How Visa
Policies Have Evolved over Time", Journal of Ethnic and Migration Studies, 41(8):
1192-1213.
McLaren, L. M. (2004) “Opposition to European integration and fear of loss of national
identity: Debunking a basic assumption regarding hostility to the integration project”.
European Journal of Political Research 43: 895-911.
McLaren, L. M. (2007) “Explaining Opposition to Turkish Membership of the EU”.
European Union Politics 8(2): 251‐278.
Nedelcu, M. (2012) "Migrants' New Transnational Habitus: Rethinking Migration Through a
Cosmopolitan Lens in the Digital Age", Journal of Ethnic and Migration Studies
38(9): 1339-1356.
Nobles, J., Rubalcava, L. and Teruel, G. (2015) “After spouses depart: emotional wellbeing
among nonmigrant Mexican mothers”, Social Science and Medicine 132: 236–244.
Nuttall, C. (2017) “Euroscepticism grows in richer Western Balkans countries”, Inellinews,
published June 14, 2017. Available online at:
http://www.intellinews.com/euroscepticism-grows-in-richer-western-balkans-
countries-125370/
Pfutze, T. (2012). “Does migration promote democratization? Evidence from the Mexican
transition.” Journal of Comparative Economics 40(2): 159–175.
42
Pottie‐Sherman, Y. & Wilkes, R. (2017) "Does Size Really Matter? On the Relationship
between Immigrant Group Size and Anti‐Immigrant Prejudice". International
Migration Review 51(1): 218-250.
Pop-Eleches, G. (2007) "Between Historical Legacies and the Promise of Western
Integration: Democratic Conditionality after Communism", East European Politics &
Societies 21(1): 142-161.
Pridham, G. (2005) Designing Democracy: EU Enlargement and Regime Change in
Postcommunist Europe. London: Palgrave Macmillan.
Ray, L. (2003) “Reconsidering the Link between Incumbent Support and Pro-EU Opinion”.
European Union Politics 4(3): 259–279.
Recchi, E. & Kuhn, T. (2013) “Europeans’ Space Sets and the Political Legitimacy of the
EU”. In Kauppi, N. (ed.): A Political Sociology of Transnational Europe. Colchester:
ECPR Press.
Sánchez-Cuenca, I. (2000) "The Political Basis of Support for European
Integration", European Union Politics 1(2): 147-171.
Schimmelfennig, F. (2007) "European Regional Organizations, Political Conditionality, and
Democratic Transformation in Eastern Europe", East European Politics & Societies
21(1): 126-141.
Schimmelfennig, F. and Sedelmeier, U. (2008) Candidate Countries and Conditionality. In:
Graziano P., Vink M.P. (Eds) Europeanization. Palgrave Macmillan, London.
43
Schuck, A.R.T. and De Vreese, C.H. (2006) "Between Risk and Opportunity: News Framing
and its Effects on Public Support for EU Enlargement", European Journal of
Communication 21(1): 5-32.
Tanasoiu, C. and Colonescu, C. (2008) “Determinants of Support for European Integration:
The Case of Bulgaria”. European Union Politics 9(3): 363-377.
Tucker, J.A., Pacek, A.C. and Berinsky, A.J. (2002) “Transitional Winners and Losers:
Attitudes toward EU Membership in Post-Communist Countries”, American Journal
of Political Science 46 (3), 557–71.
Verhaegen, S., Hooghe, M., and Quintelier, E. (2014). “European identity and support for
European integration: A matter of perceived economic benefits?” Kyklos 67: 295-
314.
Vliegenthart, R., Schuck, A.R.T., Boomgaarden, H.G. and De Vreese, C.H. (2008) "News
coverage and support for European integration, 1990-2006", International Journal of
Public Opinion Research 20(4): 415-439.
Yamron, E. (2017) “Serbian Subversion: How the European Union is Failing in the Balkans”,
Brown Political Review, published on May 11, 2017. Available online at
http://www.brownpoliticalreview.org/2017/05/serbian-subversion-european-union-
failing-balkans/
Zimmerman, W. (1987) Open Borders, Nonalignment, and the Political Evolution of
Yugoslavia. Princeton: Princeton University Press.
1
Supplementary Appendix for the article: “To Europe or not to Europe? Migration and public support for joining the European Union in the Western Balkans”
This document contains:
• Summary statistics about the variables included in the analysis (Table A1, page 2) • Complete econometric output for Tables 1, 2 and 3 (Tables A2, A3 and A4 on page 3, 5
and 7, respectively).
2
Table A1: Summary statistics of the variables included in the analysis
Mean Std. Dev. Min Max Voting yes in a hypothetical EU referendum 0.839 0.368 0 1 EU membership good thing 2.593 0.686 1 3 Well informed about the EU 2.572 0.760 1 4 Intentions to move (EU) 0.145 0.352 0 1 Intentions to move (non-EU) 0.074 0.263 0 1 Relatives abroad (EU) 0.256 0.436 0 1 Relatives abroad (non-EU) 0.102 0.303 0 1 Return migrant (EU) 0.064 0.245 0 1 Return migrant (non-EU) 0.033 0.179 0 1 Age 41.600 16.425 15 99 Female 0.499 0.500 0 1 Primary education 0.304 0.460 0 1 Secondary education 0.537 0.499 0 1 Tertiary education 0.159 0.366 0 1 1st income quintile 0.169 0.375 0 1 2nd income quintile 0.172 0.378 0 1 3rd income quintile 0.165 0.371 0 1 4th income quintile 0.163 0.369 0 1 5th income quintile 0.155 0.362 0 1 Income: no answer 0.176 0.381 0 1 Standard of living worse 2.094 0.814 1 3 Unemployed 0.130 0.336 0 1 Orthodox 0.382 0.486 0 1 Catholic 0.171 0.376 0 1 Muslim 0.401 0.490 0 1 Religion: other 0.046 0.210 0 1 Support for national government 2.000 0.905 1 4 Strong European identity 3.031 1.263 1 5 Helpful international community 3.449 0.970 1 5 Year 2009 0.322 0.467 0 1 Year 2010 0.343 0.475 0 1 Year 2011 0.335 0.472 0 1 Albania 0.156 0.363 0 1 Bosnia and Herzegovina 0.142 0.349 0 1 Croatia 0.131 0.338 0 1 Macedonia 0.141 0.348 0 1 Montenegro 0.131 0.338 0 1 Kosovo 0.157 0.364 0 1 Serbia 0.140 0.347 0 1
Source: Gallup Balkan Monitor and authors’ calculations.
3
Table A2: Complete econometric output for Table 1 (this table reports probit coefficients and not the marginal effects)
A. Dependent variable:
Voting yes in a hypothetical EU referendum; binary probit coefficients
B. Dependent variable: EU membership a good thing;
ordered probit coefficients (1) (2) (3) (4) (5) (6) (7) (8) Intentions to move to the EU 0.153*** - - 0.128** 0.067 - - 0.051 (0.055) - - (0.056) (0.043) - - (0.044) Relatives in the EU - 0.128*** - 0.109*** - 0.079** - 0.070** - (0.041) - (0.041) - (0.032) - (0.032) EU return migrant - - 0.204*** 0.162** - - 0.097 0.074 - - (0.072) (0.073) - - (0.063) (0.064) Age -0.015*** -0.016*** -0.016*** -0.015*** -0.014*** -0.014*** -0.014*** -0.014*** (0.005) (0.005) (0.005) (0.005) (0.004) (0.004) (0.004) (0.004) Age squared/100 0.017*** 0.017*** 0.018*** 0.017*** 0.014*** 0.014*** 0.014*** 0.014*** (0.006) (0.006) (0.006) (0.006) (0.004) (0.004) (0.004) (0.004) Female 0.005 0.003 0.011 0.010 0.003 0.002 0.006 0.005 (0.032) (0.032) (0.033) (0.033) (0.025) (0.025) (0.025) (0.025) Primary education 0.029 0.027 0.027 0.029 0.034 0.034 0.034 0.034 (0.046) (0.046) (0.046) (0.046) (0.036) (0.036) (0.036) (0.036) Tertiary education 0.073* 0.072* 0.074* 0.075* 0.056* 0.056* 0.057* 0.057* (0.042) (0.042) (0.042) (0.042) (0.032) (0.031) (0.031) (0.032) 2nd income quintile -0.033 -0.038 -0.034 -0.034 -0.023 -0.026 -0.024 -0.025 (0.060) (0.060) (0.059) (0.060) (0.046) (0.046) (0.046) (0.046) 3rd income quintile -0.114** -0.116** -0.114** -0.115** -0.088* -0.090* -0.089* -0.090* (0.057) (0.057) (0.057) (0.057) (0.046) (0.046) (0.046) (0.046) 4th income quintile -0.012 -0.022 -0.019 -0.019 -0.015 -0.020 -0.018 -0.019 (0.058) (0.058) (0.058) (0.058) (0.046) (0.046) (0.046) (0.046) 5th income quintile -0.080 -0.090 -0.089 -0.088 0.004 -0.003 -0.000 -0.002 (0.062) (0.062) (0.062) (0.062) (0.049) (0.049) (0.049) (0.049) Income: no answer -0.109* -0.113** -0.117** -0.110* -0.130*** -0.131*** -0.133*** -0.131*** (0.056) (0.056) (0.056) (0.056) (0.045) (0.045) (0.045) (0.045) Standard of living worse -0.115*** -0.108*** -0.110*** -0.112*** -0.086*** -0.082*** -0.084*** -0.084*** (0.023) (0.023) (0.023) (0.023) (0.018) (0.018) (0.018) (0.018) Unemployed -0.046 -0.045 -0.046 -0.045 0.013 0.013 0.012 0.013 (0.054) (0.054) (0.054) (0.053) (0.041) (0.041) (0.041) (0.041) Catholic 0.051 0.048 0.049 0.049 0.102** 0.100** 0.101** 0.098** (0.060) (0.060) (0.060) (0.060) (0.048) (0.048) (0.048) (0.048) Muslim 0.469*** 0.462*** 0.472*** 0.458*** 0.481*** 0.476*** 0.481*** 0.475*** (0.059) (0.058) (0.059) (0.059) (0.043) (0.043) (0.043) (0.043) Religion: other 0.248*** 0.249*** 0.245*** 0.244*** 0.115** 0.113** 0.113** 0.112** (0.077) (0.077) (0.077) (0.077) (0.052) (0.052) (0.052) (0.052) Support for national government 0.096*** 0.094*** 0.094*** 0.095*** 0.100*** 0.099*** 0.098*** 0.099*** (0.022) (0.022) (0.022) (0.022) (0.018) (0.018) (0.018) (0.018) Well informed about the EU 0.078*** 0.077*** 0.075*** 0.075*** 0.210*** 0.209*** 0.208*** 0.207*** (0.023) (0.023) (0.023) (0.023) (0.019) (0.019) (0.019) (0.019) Strong European identity 0.193*** 0.192*** 0.192*** 0.192*** 0.148*** 0.147*** 0.147*** 0.147*** (0.015) (0.015) (0.015) (0.015) (0.011) (0.011) (0.011) (0.011) Helpful international community 0.301*** 0.299*** 0.300*** 0.302*** 0.269*** 0.268*** 0.269*** 0.269*** (0.019) (0.019) (0.019) (0.019) (0.015) (0.015) (0.015) (0.015) Albania 0.414*** 0.415*** 0.420*** 0.388*** 0.476*** 0.474*** 0.479*** 0.460*** (0.082) (0.082) (0.081) (0.083) (0.059) (0.059) (0.058) (0.059) Bosnia and Herzegovina 0.145** 0.145** 0.148** 0.145** 0.133*** 0.133*** 0.134*** 0.133*** (0.059) (0.059) (0.059) (0.059) (0.046) (0.046) (0.046) (0.046) Croatia -0.861*** -0.865*** -0.868*** -0.864*** -0.636*** -0.637*** -0.640*** -0.635*** (0.065) (0.065) (0.065) (0.065) (0.051) (0.051) (0.051) (0.051) FYR Macedonia 0.006 0.019 0.013 0.009 -0.038 -0.032 -0.034 -0.035
4
(0.059) (0.059) (0.058) (0.059) (0.043) (0.043) (0.043) (0.043) Montenegro 0.226*** 0.236*** 0.229*** 0.236*** 0.280*** 0.285*** 0.281*** 0.286*** (0.065) (0.065) (0.065) (0.065) (0.046) (0.046) (0.046) (0.046) Kosovo 0.050 0.061 0.060 0.054 0.445*** 0.449*** 0.450*** 0.446*** (0.071) (0.071) (0.071) (0.071) (0.058) (0.058) (0.058) (0.058) Year 2010 -0.079** -0.069* -0.076** -0.077** -0.139*** -0.134*** -0.137*** -0.138*** (0.039) (0.039) (0.039) (0.039) (0.030) (0.030) (0.030) (0.030) Year 2011 -0.238*** -0.224*** -0.231*** -0.231*** -0.209*** -0.203*** -0.207*** -0.205*** (0.040) (0.040) (0.040) (0.040) (0.030) (0.031) (0.030) (0.031) Constant -0.285* -0.274 -0.234 -0.295* - - - - (0.172) (0.171) (0.171) (0.171) - - - -
cut1 - - - - 0.234* 0.231* 0.210 0.236* - - - - (0.131) (0.130) (0.130) (0.131) cut2 - - - - 1.232*** 1.230*** 1.208*** 1.235*** - - - - (0.134) (0.134) (0.134) (0.135)
Observations 16858 16858 16858 16858 17871 17871 17871 17871 Pseudo R2 0.257 0.257 0.257 0.258 0.191 0.191 0.191 0.191 Chi2 2084.307 2084.613 2081.838 2124.587 2761.454 2760.594 2760.812 2762.248 p > Chi2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Notes: *** p<0.01, ** p<0.05, * p<0.1, robust standard errors in parentheses. Reference groups: secondary education, 1st income quintile, Orthodox, living in Serbia, interviewed in year 2009.
5
Table A3: Complete econometric output for Table 2 (this table reports probit coefficients and not the marginal effects)
Dependent variable: Well informed about the EU;
ordered probit coefficients
(1) (2) (3) (4) Intentions to move to the EU -0.031 - - -0.056*
(0.032) - - (0.032) Relatives in the EU - 0.029 - 0.019
- (0.024) - (0.025) EU return migrant - - 0.226*** 0.233***
- - (0.043) (0.043) Age 0.012*** 0.012*** 0.011*** 0.011***
(0.003) (0.003) (0.003) (0.003) Age squared/100 -0.014*** -0.014*** -0.013*** -0.013***
(0.004) (0.004) (0.004) (0.004) Female -0.329*** -0.329*** -0.318*** -0.319***
(0.021) (0.021) (0.021) (0.021) Primary education -0.345*** -0.345*** -0.345*** -0.346***
(0.028) (0.028) (0.028) (0.028) Tertiary education 0.288*** 0.288*** 0.291*** 0.291***
(0.025) (0.025) (0.025) (0.025) 2nd income quintile 0.008 0.007 0.008 0.007
(0.037) (0.037) (0.037) (0.037) 3rd income quintile 0.020 0.019 0.020 0.019
(0.036) (0.036) (0.036) (0.036) 4th income quintile 0.111*** 0.110*** 0.109*** 0.106***
(0.037) (0.037) (0.036) (0.036) 5th income quintile 0.165*** 0.163*** 0.158*** 0.155***
(0.040) (0.039) (0.040) (0.040) Income: no answer 0.034 0.035 0.032 0.030
(0.037) (0.037) (0.037) (0.037) Standard of living worse -0.063*** -0.064*** -0.064*** -0.062***
(0.014) (0.014) (0.014) (0.014) Unemployed -0.008 -0.008 -0.009 -0.009
(0.034) (0.034) (0.034) (0.034) Catholic 0.147*** 0.145*** 0.144*** 0.143***
(0.041) (0.041) (0.041) (0.041) Muslim 0.116*** 0.114*** 0.112*** 0.111***
(0.031) (0.031) (0.031) (0.031) Religion: other 0.174*** 0.172*** 0.170*** 0.170***
(0.050) (0.050) (0.050) (0.050) Support for national government 0.061*** 0.062*** 0.061*** 0.060***
(0.013) (0.013) (0.013) (0.013)
6
Strong European identity 0.120*** 0.120*** 0.119*** 0.119***
(0.009) (0.009) (0.009) (0.009) Helpful international community 0.061*** 0.061*** 0.061*** 0.061***
(0.013) (0.013) (0.013) (0.013) Albania -0.135*** -0.144*** -0.160*** -0.154***
(0.041) (0.041) (0.040) (0.041) Bosnia and Herzegovina 0.134*** 0.134*** 0.135*** 0.135***
(0.038) (0.038) (0.038) (0.038) Croatia -0.036 -0.032 -0.035 -0.038
(0.046) (0.046) (0.046) (0.046) FYR Macedonia 0.004 0.004 0.002 0.006
(0.037) (0.037) (0.037) (0.037) Montenegro -0.169*** -0.167*** -0.163*** -0.163***
(0.041) (0.041) (0.041) (0.041) Kosovo 0.059 0.057 0.060 0.063
(0.045) (0.045) (0.045) (0.045) Year 2010 0.050** 0.049** 0.046* 0.048**
(0.024) (0.024) (0.024) (0.024) Year 2011 -0.096*** -0.096*** -0.097*** -0.094***
(0.025) (0.025) (0.025) (0.025) cut1 -0.911*** -0.898*** -0.917*** -0.929***
(0.102) (0.101) (0.101) (0.102) cut2 0.594*** 0.607*** 0.590*** 0.578***
(0.102) (0.101) (0.101) (0.102) cut3 2.030*** 2.044*** 2.029*** 2.017***
(0.103) (0.102) (0.102) (0.103) Observations 18539 18539 18539 18539 Pseudo R2 0.055 0.055 0.056 0.056 Chi2 1363.185 1365.542 1396.015 1398.432 p > Chi2 0.000 0.000 0.000 0.000
Notes: *** p<0.01, ** p<0.05, * p<0.1, robust standard errors in parentheses. Reference groups: secondary education, 1st income quintile, Orthodox, living in Serbia, interviewed in year 2009.
7
Table A4: Complete econometric output for Table 3 (this table reports probit coefficients and not the marginal effects)
A. Dependent variable: Voting yes in a hypothetical EU referendum;
binary probit coefficients
B. Dependent variable: EU membership a good thing;
ordered probit coefficients
(1) (2) (3) (4) (5) (6) (7) (8)
Intentions to move to the EU 0.170*** - - 0.144** 0.082* - - 0.066
(0.056) - - (0.056) (0.044) - - (0.044) Relatives in the EU - 0.133*** - 0.111*** - 0.072** - 0.061*
- (0.041) - (0.042) - (0.033) - (0.033) EU return migrant - - 0.200*** 0.156** - - 0.094 0.071
- - (0.072) (0.073) - - (0.063) (0.064) Intentions to move (non-EU) 0.141** - - 0.137** 0.129*** - - 0.136***
(0.063) - - (0.063) (0.047) - - (0.047) Relatives abroad (non-EU) - 0.041 - 0.036 - -0.050 - -0.056
- (0.051) - (0.052) - (0.038) - (0.039) Return migrant (non-EU) - - -0.094 -0.115 - - -0.069 -0.077
- - (0.092) (0.092) - - (0.067) (0.068) Age -0.014*** -0.016*** -0.016*** -0.014*** -0.013*** -0.014*** -0.014*** -0.013***
(0.005) (0.005) (0.005) (0.005) (0.004) (0.004) (0.004) (0.004) Age squared/100 0.017*** 0.017*** 0.017*** 0.017*** 0.014*** 0.014*** 0.014*** 0.014***
(0.006) (0.006) (0.006) (0.006) (0.004) (0.004) (0.004) (0.004) Female 0.006 0.002 0.009 0.008 0.005 0.002 0.005 0.006
(0.032) (0.032) (0.033) (0.033) (0.025) (0.025) (0.025) (0.025) Primary education 0.029 0.028 0.027 0.030 0.035 0.033 0.034 0.033
(0.046) (0.046) (0.046) (0.046) (0.036) (0.036) (0.036) (0.036) Tertiary education 0.072* 0.071* 0.077* 0.076* 0.056* 0.057* 0.059* 0.059*
(0.042) (0.042) (0.042) (0.042) (0.032) (0.032) (0.032) (0.032) 2nd income quintile -0.032 -0.038 -0.034 -0.033 -0.022 -0.025 -0.023 -0.022
(0.059) (0.060) (0.059) (0.059) (0.046) (0.046) (0.046) (0.046) 3rd income quintile -0.116** -0.117** -0.114** -0.118** -0.089* -0.089* -0.089* -0.090*
(0.057) (0.057) (0.057) (0.057) (0.046) (0.046) (0.046) (0.046) 4th income quintile -0.011 -0.023 -0.020 -0.019 -0.014 -0.019 -0.018 -0.016
(0.058) (0.058) (0.058) (0.058) (0.046) (0.046) (0.046) (0.046) 5th income quintile -0.080 -0.092 -0.088 -0.088 0.004 -0.001 0.000 0.001
(0.061) (0.062) (0.062) (0.062) (0.049) (0.049) (0.049) (0.049) Income: no answer -0.108* -0.114** -0.117** -0.110* -0.129*** -0.129*** -0.133*** -0.128***
(0.056) (0.056) (0.056) (0.056) (0.045) (0.045) (0.045) (0.045) Standard of living worse -0.117*** -0.108*** -0.110*** -0.114*** -0.087*** -0.083*** -0.084*** -0.086***
(0.023) (0.023) (0.023) (0.023) (0.018) (0.018) (0.018) (0.018) Unemployed -0.050 -0.044 -0.045 -0.046 0.010 0.012 0.013 0.011
(0.053) (0.053) (0.053) (0.053) (0.040) (0.041) (0.041) (0.041) Catholic 0.055 0.049 0.048 0.053 0.106** 0.099** 0.100** 0.100**
(0.060) (0.060) (0.060) (0.060) (0.048) (0.048) (0.048) (0.048) Muslim 0.473*** 0.464*** 0.472*** 0.463*** 0.485*** 0.475*** 0.481*** 0.476***
(0.059) (0.058) (0.059) (0.058) (0.043) (0.043) (0.043) (0.043) Religion: other 0.244*** 0.249*** 0.245*** 0.241*** 0.113** 0.113** 0.113** 0.110**
(0.077) (0.077) (0.077) (0.077) (0.052) (0.052) (0.052) (0.052) Support for national government 0.099*** 0.094*** 0.094*** 0.098*** 0.103*** 0.099*** 0.098*** 0.102***
(0.022) (0.022) (0.022) (0.022) (0.018) (0.018) (0.018) (0.018) Well informed about the EU 0.077*** 0.077*** 0.076*** 0.075*** 0.209*** 0.209*** 0.209*** 0.208***
(0.023) (0.024) (0.023) (0.023) (0.019) (0.019) (0.019) (0.019) Strong European identity 0.193*** 0.192*** 0.192*** 0.193*** 0.148*** 0.147*** 0.148*** 0.148***
(0.015) (0.015) (0.015) (0.015) (0.011) (0.011) (0.011) (0.011) Helpful international community 0.301*** 0.299*** 0.300*** 0.302*** 0.269*** 0.269*** 0.269*** 0.269***
(0.019) (0.019) (0.019) (0.019) (0.015) (0.015) (0.015) (0.015) Albania 0.403*** 0.414*** 0.421*** 0.379*** 0.467*** 0.475*** 0.480*** 0.453***
8
(0.082) (0.082) (0.081) (0.083) (0.059) (0.059) (0.058) (0.059) Bosnia and Herzegovina 0.145** 0.142** 0.150** 0.144** 0.133*** 0.136*** 0.135*** 0.138***
(0.059) (0.059) (0.059) (0.059) (0.046) (0.046) (0.046) (0.046) Croatia -0.861*** -0.866*** -0.868*** -0.864*** -0.636*** -0.636*** -0.640*** -0.634***
(0.065) (0.065) (0.065) (0.065) (0.051) (0.051) (0.051) (0.051) FYR Macedonia -0.006 0.014 0.016 -0.004 -0.049 -0.026 -0.033 -0.039
(0.059) (0.059) (0.058) (0.059) (0.043) (0.043) (0.043) (0.043) Montenegro 0.220*** 0.234*** 0.232*** 0.232*** 0.275*** 0.287*** 0.283*** 0.286***
(0.065) (0.065) (0.065) (0.065) (0.046) (0.046) (0.046) (0.046) Kosovo 0.043 0.058 0.061 0.046 0.438*** 0.453*** 0.451*** 0.443***
(0.071) (0.071) (0.071) (0.071) (0.058) (0.058) (0.058) (0.058) Year 2010 -0.075* -0.068* -0.077** -0.073* -0.135*** -0.135*** -0.138*** -0.135***
(0.039) (0.039) (0.039) (0.039) (0.030) (0.030) (0.030) (0.030) Year 2011 -0.239*** -0.224*** -0.231*** -0.232*** -0.210*** -0.203*** -0.207*** -0.206***
(0.040) (0.040) (0.040) (0.040) (0.030) (0.031) (0.030) (0.031) Constant -0.320* -0.281* -0.238 -0.339** - - - -
(0.173) (0.170) (0.170) (0.171) - - - - cut1 - - - - 0.267** 0.224* 0.211 0.264**
- - - - (0.132) (0.130) (0.130) (0.132) cut2 - - - - 1.266*** 1.223*** 1.210*** 1.263***
- - - - (0.135) (0.134) (0.134) (0.135) Observations 16858 16858 16858 16858 17871 17871 17871 17871 Pseudo R2 0.257 0.257 0.257 0.259 0.191 0.191 0.191 0.191 Chi2 2093.338 2087.684 2079.904 2134.609 2777.588 2759.721 2758.910 2776.456 p > Chi2 0.000 0.000 0.000 0.000 0.000 0.000 0.000 0.000
Notes: *** p<0.01, ** p<0.05, * p<0.1, robust standard errors in parentheses. Reference groups: secondary education, 1st income quintile, Orthodox, living in Serbia, interviewed in year 2009.